A slow hollow-fiber filter is usually clogged, not dead, and a backflush with clean water restores most of its flow rate. The failures that actually end a filter's life happen after the backflush: freezing, wet storage, expired chemical purifiers. None of them leave a visible sign, which is why the next trip is where you find out.
Put your hands where mine were on the Tuesday after a four-day loop. Kitchen counter, cutting board shoved to the side, a one-liter bottle, the filter, and a phone propped against the backsplash with the stopwatch running. On the last morning out, that filter had taken a little over ten minutes to fill the same bottle. On day one it took about ninety seconds. I stood at that creek watching the stream thin to a thread and did the math on how much water I still had to carry for the climb out.
The backflush itself is the easy half
Backflushing means forcing clean water through the filter in the opposite direction: in through the clean outlet, out through the dirty inlet. The silt and biofilm that packed onto the outside of the hollow fibers gets pushed off rather than driven deeper.
How you do it depends entirely on what you are holding. Some filters ship with a syringe and a hose coupling. Gravity and squeeze systems usually have you reverse the rig, fill a clean bag, and let gravity push water backward through the element. Pumps often have a backwash adapter or a specific lever position. Read your own manual for the direction and the stroke count, because pushing water the wrong way is not the same as pushing it the right way.
The first push of the plunger tells you how loaded the filter is. On a bad one, the plunger moves maybe a quarter inch and stops, like compressing a sponge that is already full. Hold it. Wait a beat. Then it gives, and what comes out the inlet side is gray-brown water that looks nothing like what you put in.
Work in batches. Ten or twenty strokes, dump the dirty water, refill with clean, repeat. I use tap water at home. On the trail I use water I have already filtered, because pushing silty creek water backward just seats the grit harder into the fibers.
Then flush forward once and time the flow again. On that Tuesday, ten minutes dropped to about two. Not back to ninety seconds; a filter rarely comes all the way back. But close enough to carry another season.
The damage a backflush cannot undo
Now the part that matters more than the procedure.
Hollow fibers are tubes with microscopic pores in the wall, and they hold water. Water expands when it freezes. A fiber that freezes can crack, and a cracked fiber is not a clogged fiber. A clog blocks flow. A crack is a hole. Water moves through it freely, sometimes faster than before, and everything the membrane was supposed to stop goes straight through with it.
You cannot see it. You cannot feel it. There is no field test that will tell you. That is why the guidance from hollow-fiber makers tends to be blunt: if it froze, treat it as compromised and replace it. I do not argue with that. The cost of being wrong is a gastrointestinal illness ten miles from a trailhead.
So the filter sleeps where you sleep. Inside the sleeping bag, in a jacket pocket against your chest, at the foot of the quilt, anywhere that stays above freezing through the night. Not the pack's outer pocket. Not the lid. Not the bear hang, and not the vestibule. In real cold, that also means filtering before bed instead of at dawn, so the filter goes into the bag wet and warm instead of frozen and useless.
The same logic applies to pressure. A syringe is gentle. A garden hose or a bike pump is not, and over-pressurizing a fiber bundle ruptures it in a way that looks exactly like freeze damage from the inside. If your manual does not call for it, do not do it.
Drying is the step everyone skips
Backflushing fixes flow. It does not fix what grows afterward.
A filter put away damp sits in a dark, warm, sealed stuff sack with a film of organic matter on it. That is a culture dish. Mold and bacteria colonize the element and the hoses, and the smell when you open the sack next spring tells you what happened. What you cannot smell is the inside of the tubing.
So shake out what you can, then hang the housing and the hoses separately in a warm room with air moving, out of direct sun. UV is hard on the plastics and the O-rings. Give it a day or two, longer in a humid house or a damp basement. I hang the hoses loose with both ends open so air moves through the middle instead of sitting in a coil. A fan helps. So does propping the housing with both openings facing down so it drains.
Storage: two schools, and your manual wins
Once it is dry, there are two storage approaches, and the right one is whichever your manufacturer printed.
Bone dry is one school. Nothing grows in a desiccated filter, and that is the whole appeal.
The other is a small amount of clean water plus a preservative, if the manual calls for one, to keep the O-rings and seals from drying out and stiffening. Rubber that has sat compressed and dry for six months can weep or crack, and a weeping seal is a pathway around the membrane. Same practical result as a cracked fiber, arrived at more slowly.
I follow the manual for the specific filter in my hand rather than a rule I made up. What I do not do is mix approaches. A filter stored wet with tap water and no preservative is the worst of both, neither dry nor protected.
Some manufacturers also print a periodic disinfection procedure, with an exact concentration and contact time. That printed page is the only place you should read about it. Getting the recipe wrong in either direction damages the membrane or leaves residue you end up drinking, so if your manual does not print one, do not invent it.
Chemical and UV purifiers age on the shelf
Not everything in the water kit is a filter, and the other two categories fail on a different clock.
Chemical purifiers, the liquid bottles and the tablets, carry expiration dates and lose potency as they age. Heat accelerates it. A bottle of tablets that spent two summers in a glovebox is not the same product that came off the shelf, even if the printed date has not passed, because that date assumes the storage conditions on the label. I write the month and year on the bottle with a marker when it goes into the bin and check it before every trip, the same way I check dates on the food I am packing.
UV purifiers are consumable in two ways. The lamp has an hour budget and the batteries have a finite number of cycles, and when either runs down, the device keeps turning on and doing less than the indicator suggests. The quartz sleeve the light passes through also picks up a film, from water, from hands, from the inside of a pack, and a dirty sleeve absorbs UV before it reaches the water. Wipe it with a soft, lint-free cloth the way the manual describes. Do not scrub it.
The five-minute shakedown
Before the next trip, run the check that tells you whether any of this worked.
Fill a bottle from the tap, run the filter, time it. Compare the number to what you remember, and if you do not trust your memory, write it on the bottle with a Sharpie. I do not trust mine.
If the flow is worse than last time, backflush again. If it is still worse, and the filter never froze, you have reached the end of what maintenance can do, and a new element or a new filter costs less than getting sick on day three.
If it froze at any point, skip the test. Flow rate only tells you about clogs. It says nothing about integrity, and integrity is the thing you were actually buying.
